Radial frequency adaptation suggests polar-based coding of local shape cues

Jason Bell, Edwin Dickinson, David Badcock

Research output: Contribution to journalArticlepeer-review

37 Citations (Web of Science)

Abstract

The study of shape processing in the human visual system has frequently employed radial frequency (RF) patterns as conveniently manipulable stimuli. This study uses an adaptation paradigm to investigate how local shape information is sampled in the processing of RF contour shapes. Experiment 1 measured thresholds for detecting a fixed mean radius RF contour following adaptation to RF patterns which, in separate conditions, varied in mean radius and radial frequency. Results reveal that, adaptation is strongly tuned for RF over a range of pattern radii, but is not tuned for the number of cycles of radial modulation per visual degree of contour length; a characteristic that changes with both radius and radial frequency. Experiment 2 manipulated the polar angle separation on the fronto-parallel plane between curvature features on a fixed RE by foreshortening the pattern appearance (consistent with a rotation in depth) and shows that RF shape processing is tuned for fronto-parallel separation angles between curvature features. Results were near identical when a stereo rotation cue was added to the perspective modified RF. In the second part of Experiment 2 we showed that RF shape adaptation is also tuned for the polar angular extent of the curvature represented by the lobe at that angle. Collectively, our results indicate that the polar angle at which local curvature features appear, in addition to the angular extent of the curvature feature at that location, are both critical parameters for coding specific RF shapes. Crown copyright (C) 2008 Published by Elsevier Ltd. All rights reserved.
Original languageEnglish
Pages (from-to)2293-2301
JournalVision Research
Volume48
DOIs
Publication statusPublished - 2008

Fingerprint

Dive into the research topics of 'Radial frequency adaptation suggests polar-based coding of local shape cues'. Together they form a unique fingerprint.

Cite this